uis-rnn VS orange

Looks like you can get an put of the box here: https://github.com/google/uis-rnn

Discussion: - When I started this, I thought I would use something like the VoxSort Diarization and it would be easy. But these apps are terrible, especially in recognizing Joey apart from Garnt. Connor has a distinct voice so it was recognizable but still bad. But I didn't think Joey's and Garnt's voices were so similar. - Tested the thing and it's accuracy is almost 99%. - You can still improve this by cutting the episode into smaller chunk but 1 second is the maximum for my computer, any smaller than that i will run out of RAM. I can work to get around this but hey I'm lazy. - The library to implement yourself from google.

I used this algorithm from Google to determine "who spoke when".

I know I've tooted its horn before, but Orange3 is a pretty neat Python-based GUI platform that makes this and a metric buttload of other statistical/ML techniques available to non-programmer types.

Just watch out for null character `x00` in the corpus. That always seems to kill it stone dead.

https://orangedatamining.com/

https://orange3.readthedocs.io/projects/orange-visual-progra...

For all you folks who aren't ace programmer types, the Orange3[1] platform gives you a very miniaturized[2] ability to turn out these sorts of visualizations very rapidly. It's not the most stable thing in the world, but the node-based ML workflow designer is worth the price of admission all by itself.

[1] https://orangedatamining.com/

[2] The Wikipedia extension in Text limits each search result to 25 articles, so sucking all of Wikipedia is . . well, Orange text analytics crashes when I look at it sideways with a null character, so let's not think about what would happen.

First is identifying the "similar" things in a corpus. Best way I know to do that, for non-programmer audiences, is the Orange Data Mining tool, which gives you a node-based text mining interface to perform statistical analysis on text. Hierarchical Clustering shows - very rapidly - how similar your "modules" are, which ones are most similar. There's many other techniques (semantic viewer, similarity hash, etc) as well - the right one will depend on how your content is laying about.

Both tools are ripoffs of a data mining framework named Orange 3

You should also check out Orange Data Mining, it allows to create a lot of charts, filter data from a chart to another, build ML models, predictions and a lot more. And you can do it with zero code.

You can start with a free GUI based tool Orange. It is a component based data science workflow tool, which you can use to handle 60-75% of the traditional data science tasks from classification, regression, to basic neural networks.